003_Project_analytics.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "18bWwaWUrFJf"
      },
      "source": [
        "Thesis: It's possible to detect delays on a project schedule for construction analisying the tasks, expected times of start, resources, task dependencies and weather conditions."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "dSd45z9-tVMm"
      },
      "source": [
        "The steps to perform the analysis are:\n",
        "\n",
        "1.   Remove tasks with non text names (1.2, 33, etc)\n",
        "2.   Categorize tasks by its function\n",
        "3.   Add resource type and amount to the task\n",
        "4.   Add weather temp to the task\n",
        "5.   Amount of dependencies per task\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "XF4bQ12bEY-r"
      },
      "source": [
        "1.  Categorize tasks by function"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "v57_dn5ViTM1"
      },
      "source": [
        "# Settings"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "jXu7osOajDQY"
      },
      "source": [
        "Connect to google drive to access the dataset of projects."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "JuUETIxv3q-L",
        "outputId": "18429a02-42bb-404f-b142-1a984d882107",
        "tags": [
          "parameters"
        ]
      },
      "outputs": [],
      "source": [
        "csv_folder_path = './content/drive/MyDrive/BILT'\n",
        "# csv_folder_path = './CSV'\n",
        "classified_tasks_filepath = './classified_tasks.csv'\n",
        "training_data_path = './content/drive/MyDrive/BILT/Training_classified_tasks.csv'\n",
        "Tasks_to_be_preditected ='./Baseline Zone 2 Rev/TASK.csv'\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7X9fv8BxigEl",
        "outputId": "c532d6f7-13b6-403a-dddf-a27f7728b27b"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Requirement already satisfied: pandas in c:\\users\\pablod\\anaconda3\\lib\\site-packages (2.2.1)\n",
            "Requirement already satisfied: numpy<2,>=1.23.2 in c:\\users\\pablod\\anaconda3\\lib\\site-packages (from pandas) (1.26.4)\n",
            "Requirement already satisfied: python-dateutil>=2.8.2 in c:\\users\\pablod\\anaconda3\\lib\\site-packages (from pandas) (2.9.0.post0)\n",
            "Requirement already satisfied: pytz>=2020.1 in c:\\users\\pablod\\anaconda3\\lib\\site-packages (from pandas) (2024.1)\n",
            "Requirement already satisfied: tzdata>=2022.7 in c:\\users\\pablod\\anaconda3\\lib\\site-packages (from pandas) (2024.1)\n",
            "Requirement already satisfied: six>=1.5 in c:\\users\\pablod\\anaconda3\\lib\\site-packages (from python-dateutil>=2.8.2->pandas) (1.16.0)\n",
            "Note: you may need to restart the kernel to use updated packages.\n",
            "Requirement already satisfied: scikit-learn in c:\\users\\pablod\\anaconda3\\lib\\site-packages (1.2.2)\n",
            "Requirement already satisfied: numpy>=1.17.3 in c:\\users\\pablod\\anaconda3\\lib\\site-packages (from scikit-learn) (1.26.4)\n",
            "Requirement already satisfied: scipy>=1.3.2 in c:\\users\\pablod\\anaconda3\\lib\\site-packages (from scikit-learn) (1.11.4)\n",
            "Requirement already satisfied: joblib>=1.1.1 in c:\\users\\pablod\\anaconda3\\lib\\site-packages (from scikit-learn) (1.2.0)\n",
            "Requirement already satisfied: threadpoolctl>=2.0.0 in c:\\users\\pablod\\anaconda3\\lib\\site-packages (from scikit-learn) (2.2.0)\n",
            "Note: you may need to restart the kernel to use updated packages.\n",
            "Requirement already satisfied: numpy in c:\\users\\pablod\\anaconda3\\lib\\site-packages (1.26.4)\n",
            "Note: you may need to restart the kernel to use updated packages.\n"
          ]
        }
      ],
      "source": [
        "%pip install pandas\n",
        "%pip install scikit-learn\n",
        "%pip install numpy\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "jmBDM2_phzEb"
      },
      "source": [
        "We are going to use ML to create a relationship between tasks and categories and subcategories that will help up define to which group a task belong.\n",
        "This list was created manually and can be edited to include more categories"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "id": "9cmaWmG2go7I"
      },
      "outputs": [],
      "source": [
        "categories_subcategories = {\n",
        "    \"Preliminary\": [\"Approval\", \"Submittal\"],\n",
        "    \"Site Preparation\": [\n",
        "        \"Clearing and Excavation\",\n",
        "        \"Site Security Setup\",\n",
        "        \"Utility Connections\",\n",
        "    ],\n",
        "    \"Foundation Construction\": [\n",
        "        \"Footing Installation\",\n",
        "        \"Foundation Concrete Pour\",\n",
        "        \"Waterproofing and Drainage\",\n",
        "        \"Footing\",\n",
        "        \"BACKFILLING\",\n",
        "        \"Foundation\",\n",
        "        \"REBAR\",\n",
        "        \"GROUNDING\",\n",
        "        \"FORMWORK\",\n",
        "    ],\n",
        "    \"Structural\": [\n",
        "        \"Steel Framing Assembly\",\n",
        "        \"Concrete Structure Formation\",\n",
        "        \"Wooden Frame Construction\",\n",
        "        \"Column\",\n",
        "        \"Columns\",\n",
        "        \"Beam\",\n",
        "        \"Slab\",\n",
        "        \"Wall\",\n",
        "        \"Beams\",\n",
        "        \"Slabs\",\n",
        "        \"BRICK\",\n",
        "        \"Walls\",\n",
        "        \"MASONRY\",\n",
        "        \"Roof Trusses\",\n",
        "    ],\n",
        "    \"MEP Installation\": [\n",
        "        \"HVAC System\",\n",
        "        \"Electrical\",\n",
        "        \"Plumbing \",\n",
        "        \"Fire Protection\",\n",
        "        \"Insulation\",\n",
        "        \"CABLES\",\n",
        "        \"DUCTS\",\n",
        "        \"PIPES\",\n",
        "        \"WIRES\",\n",
        "        \"CONDUITS\",\n",
        "        \"PUMPS\",\n",
        "        \"VALVES\",\n",
        "        \"TELECOMMUNICATIONS\",\n",
        "        \"CONTROLS\",\n",
        "        \"AIR HANDLING UNIT\",\n",
        "        \"CHILLERS\",\n",
        "        \"COOLING TOWER\",\n",
        "        \"FAN COIL UNIT\",\n",
        "        \"FIRE ALARM\",\n",
        "        \"FIRE SPRINKLER\",\n",
        "        \"GENERATOR\",\n",
        "        \"MECHNICAL EQUIPMENT\",\n",
        "        \"PLUMBING FIXTURES\",\n",
        "        \"LIGHTS\",\n",
        "        \"WIRING\",\n",
        "        \"GAS\"\n",
        "\n",
        "\n",
        "    ],\n",
        "    \"Interior Finishing\": [\n",
        "        \"Drywall Installation\",\n",
        "        \"Flooring\",\n",
        "        \"Paint\",\n",
        "        \"PAINTING\",\n",
        "        \"Plaster\",\n",
        "        \"Ceiling\",\n",
        "        \"PLASTERING\",\n",
        "        \"Fixtures and Fittings\",\n",
        "        \"TILE\",\n",
        "        \"CASEWORKS\"\n",
        "    ],\n",
        "    \"Exterior Finishing\": [\n",
        "        \"Roofing\",\n",
        "        \"Siding\",\n",
        "        \"Windows\",\n",
        "        \"Doors\",\n",
        "        \"Decking\",\n",
        "        \"Paving\",\n",
        "    ],\n",
        "    \"Shipping\": [\"Shipping\", \"Receiving\", \"Delivery\", \"Transportation\", \"Manufacturing\"],\n",
        "    \"Landscaping\": [\"Soil Preparation\", \"Planting\", \"Hardscape Features\"],\n",
        "    \"Other\": [\"Other\", \"Unknown\", \"Empty\", \"NAN\", \"   \"],\n",
        "}\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "aLUBEgucJY74"
      },
      "source": [
        "This functions will travel through all the Task CSV files in the dataset collecting their names"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_St4VTa99jPk",
        "outputId": "00ee2352-fc7e-4c22-cd65-02a7ab013779"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "c:\\Users\\PabloD\\anaconda3\\Lib\\site-packages\\pandas\\core\\arrays\\masked.py:60: UserWarning: Pandas requires version '1.3.6' or newer of 'bottleneck' (version '1.3.5' currently installed).\n",
            "  from pandas.core import (\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:43: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.\n",
            "  df[col] = pd.to_datetime(df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:43: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.\n",
            "  df[col] = pd.to_datetime(df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:43: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.\n",
            "  df[col] = pd.to_datetime(df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:43: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.\n",
            "  df[col] = pd.to_datetime(df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:43: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.\n",
            "  df[col] = pd.to_datetime(df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:43: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.\n",
            "  df[col] = pd.to_datetime(df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:56: DtypeWarning: Columns (33,35) have mixed types. Specify dtype option on import or set low_memory=False.\n",
            "  task_rsrc_df = fix_data(pd.read_csv(folder_path + \"TASKRSRC.csv\"))\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:43: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.\n",
            "  df[col] = pd.to_datetime(df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:43: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.\n",
            "  df[col] = pd.to_datetime(df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:43: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.\n",
            "  df[col] = pd.to_datetime(df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:43: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.\n",
            "  df[col] = pd.to_datetime(df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:56: DtypeWarning: Columns (35,36) have mixed types. Specify dtype option on import or set low_memory=False.\n",
            "  task_rsrc_df = fix_data(pd.read_csv(folder_path + \"TASKRSRC.csv\"))\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:43: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.\n",
            "  df[col] = pd.to_datetime(df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:43: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.\n",
            "  df[col] = pd.to_datetime(df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:43: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.\n",
            "  df[col] = pd.to_datetime(df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:51: DtypeWarning: Columns (28,43) have mixed types. Specify dtype option on import or set low_memory=False.\n",
            "  task_df = fix_data(pd.read_csv(folder_path + \"TASK.csv\"))\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:43: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.\n",
            "  df[col] = pd.to_datetime(df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:43: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.\n",
            "  df[col] = pd.to_datetime(df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:43: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.\n",
            "  df[col] = pd.to_datetime(df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:43: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.\n",
            "  df[col] = pd.to_datetime(df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:43: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.\n",
            "  df[col] = pd.to_datetime(df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:43: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.\n",
            "  df[col] = pd.to_datetime(df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:43: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.\n",
            "  df[col] = pd.to_datetime(df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:43: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.\n",
            "  df[col] = pd.to_datetime(df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:43: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.\n",
            "  df[col] = pd.to_datetime(df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:43: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.\n",
            "  df[col] = pd.to_datetime(df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:43: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.\n",
            "  df[col] = pd.to_datetime(df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:44: FutureWarning: DataFrame.fillna with 'method' is deprecated and will raise in a future version. Use obj.ffill() or obj.bfill() instead.\n",
            "  df.fillna(method=\"ffill\", inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3982789888.py:94: FutureWarning: The behavior of DataFrame concatenation with empty or all-NA entries is deprecated. In a future version, this will no longer exclude empty or all-NA columns when determining the result dtypes. To retain the old behavior, exclude the relevant entries before the concat operation.\n",
            "  final_df = pd.concat(dfs, ignore_index=True)\n"
          ]
        }
      ],
      "source": [
        "import os\n",
        "import hashlib\n",
        "import pandas as pd\n",
        "\n",
        "def get_all_data_from_csvs(start_path: str = \".\",file_name:str = \"TASK\"):\n",
        "    \"\"\"Get all data from TASK.csv files in the specified directory and its subdirectories.\"\"\"\n",
        "    task_csv_files = find_task_csv_files(start_path,file_name)\n",
        "    all_data = []\n",
        "    for file_path in task_csv_files:\n",
        "        # try:\n",
        "        df = pd.read_csv(file_path)\n",
        "        # Hash the project_name to anonymize the data\n",
        "        project_name = file_path.split(\"/\")[-2] #TODO data\n",
        "        hashed_project_name = hashlib.sha1(project_name.encode()).hexdigest()\n",
        "        df['file_name'] = hashed_project_name\n",
        "        if 'task_name' in df.columns:\n",
        "            df= df[df[\"task_name\"] != '']\n",
        "        all_data.append(df)\n",
        "    df = pd.concat(all_data, ignore_index=True)\n",
        "    return df\n",
        "\n",
        "\n",
        "def find_task_csv_files(start_path: str, file_name:str):\n",
        "    \"\"\"Search for .csv files in subdirectories of the given start path.\"\"\"\n",
        "    for root, dirs, files in os.walk(start_path):\n",
        "        for file in files:\n",
        "            if file == file_name + \".csv\":\n",
        "                yield os.path.join(root, file)\n",
        "\n",
        "\n",
        "def find_task_csv_files(start_path: str):\n",
        "    \"\"\"Search for TASK.csv files in subdirectories of the given start path.\"\"\"\n",
        "    for root, dirs, files in os.walk(start_path):\n",
        "        for file in files:\n",
        "            if file == \"TASK.csv\":\n",
        "                yield root + \"/\"\n",
        "\n",
        "\n",
        "def fix_data(df):\n",
        "    \"\"\"Convert date columns to datetime format.\"\"\"\n",
        "    date_columns = [x for x in df.columns if \"date\" in x]\n",
        "    for col in date_columns:\n",
        "        df[col] = pd.to_datetime(df[col], errors=\"coerce\")\n",
        "    df.fillna(method=\"ffill\", inplace=True)\n",
        "    return df\n",
        "\n",
        "\n",
        "def read_all_csvs(folder_path):\n",
        "    files = os.listdir(folder_path)\n",
        "    project_df = fix_data(pd.read_csv(folder_path + \"PROJECT.csv\"))\n",
        "    task_df = fix_data(pd.read_csv(folder_path + \"TASK.csv\"))\n",
        "    # task_pred_df = fix_data(pd.read_csv(folder_path + \"TASKPRED.csv\"))\n",
        "    task_rsrc_df = pd.DataFrame(columns=[\"task_id\", \"rsrc_id\"])\n",
        "    rsrc_df = pd.DataFrame(columns=[\"rsrc_id\", \"rsrc_name\"])\n",
        "    if \"TASKRSRC.csv\"  in files:\n",
        "        task_rsrc_df = fix_data(pd.read_csv(folder_path + \"TASKRSRC.csv\"))\n",
        "\n",
        "    if \"RSRC.csv\"  in files:\n",
        "        rsrc_df = fix_data(pd.read_csv(folder_path + \"RSRC.csv\"))\n",
        "    # rsrc_rate_df = fix_data(pd.read_csv(folder_path + \"RSRCRATE.csv\"))\n",
        "    if 'TASKPRED.csv' in files:\n",
        "        task_pred_df = fix_data(pd.read_csv(folder_path + \"TASKPRED.csv\"))\n",
        "        pred = task_pred_df.groupby('pred_task_id').agg(task_id_count=('task_id', 'count'))\n",
        "        pred.rename(columns={'task_id_count': 'successors'}, inplace=True)\n",
        "        pred = pred.rename_axis('task_id')\n",
        "        task_df = pd.merge(task_df, pred, on=\"task_id\", how=\"left\")\n",
        "    # Merging project and tasks data on a common key\n",
        "    project_task_df = pd.merge(project_df, task_df, on=\"proj_id\", how=\"inner\")\n",
        "\n",
        "    # Merging task resources data\n",
        "    task_resources_df = pd.merge(task_rsrc_df, rsrc_df, on=\"rsrc_id\", how=\"left\")\n",
        "\n",
        "    project_task_df[\"actual_duration\"] = (\n",
        "        project_task_df[\"act_end_date\"] - project_task_df[\"act_start_date\"]\n",
        "    ).dt.days\n",
        "    project_task_df[\"late_duration\"] = (\n",
        "        project_task_df[\"late_end_date\"] - project_task_df[\"late_start_date\"]\n",
        "    ).dt.days\n",
        "    project_task_df[\"target_duration\"] = (\n",
        "        project_task_df[\"target_end_date\"] - project_task_df[\"target_start_date\"]\n",
        "    ).dt.days\n",
        "\n",
        "    final_df = pd.merge(project_task_df, task_resources_df, on=\"task_id\", how=\"left\")\n",
        "\n",
        "\n",
        "    return final_df\n",
        "\n",
        "\n",
        "folders = find_task_csv_files(csv_folder_path)\n",
        "dfs = []\n",
        "for folder in folders:\n",
        "    df = read_all_csvs(folder)\n",
        "    dfs.append(df)\n",
        "final_df = pd.concat(dfs, ignore_index=True)\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Lgo8NdaNmBA4"
      },
      "source": [
        "We merge all data from Task on all projects. We use a feature extraction to classify each task"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PlCmvPJoJY75",
        "outputId": "59c1d1f7-ab19-43c4-a555-62906c352953"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "0         0.0\n",
              "1         1.0\n",
              "2         1.0\n",
              "3         1.0\n",
              "4         2.0\n",
              "         ... \n",
              "386593    1.0\n",
              "386594    1.0\n",
              "386595    1.0\n",
              "386596    0.0\n",
              "386597    0.0\n",
              "Name: successors, Length: 386598, dtype: float64"
            ]
          },
          "execution_count": 5,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "final_df['successors'] = final_df['successors'].fillna(0)\n",
        "final_df['successors']"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "id": "zOUzELjeckYd"
      },
      "outputs": [],
      "source": [
        "\n",
        "all_tasks_names = pd.DataFrame()\n",
        "all_tasks_names[\"task_name\"]= final_df[\"task_name\"].dropna().unique()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5GrhWwtvJY75"
      },
      "source": [
        "We are going to do the same with the predecessors."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ky6qKRXkJY75"
      },
      "source": [
        "Group by 'task_pred_id' and count 'pred_task_id' which is the amount of dependencies a task has"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "P_rtaWJTJY75"
      },
      "source": [
        "## Classification\n",
        "To be able to compare different tasks we need to classify them. Given this is a demo of the project using public data, the classification will be done using text vectorization and ML classification."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GjRmr4S3JY75"
      },
      "source": [
        "First we train a model with labeled data and store the model to classify the whole dataset"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 23,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Q0wviCB_0wcm",
        "outputId": "ada6377b-b12d-4fe5-9dbe-63ac67b45579"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "                                          precision    recall  f1-score   support\n",
            "\n",
            "                Exterior Finishing_Doors       1.00      1.00      1.00         2\n",
            "              Exterior Finishing_Windows       1.00      1.00      1.00         7\n",
            "      Foundation Construction_Foundation       0.00      0.00      0.00         2\n",
            "              Interior Finishing_Ceiling       1.00      0.83      0.91         6\n",
            "                Interior Finishing_Paint       1.00      0.86      0.92         7\n",
            "              Interior Finishing_Plaster       1.00      1.00      1.00         1\n",
            "             MEP Installation_Electrical       1.00      1.00      1.00         4\n",
            "            MEP Installation_HVAC System       1.00      0.33      0.50         3\n",
            "             MEP Installation_Insulation       0.00      0.00      0.00         1\n",
            "                  MEP Installation_PUMPS       0.00      0.00      0.00         1\n",
            "              MEP Installation_Plumbing        0.62      1.00      0.77         5\n",
            "                    Preliminary_Approval       0.81      0.89      0.85        28\n",
            "                   Preliminary_Submittal       0.60      1.00      0.75         3\n",
            "Site Preparation_Clearing and Excavation       0.00      0.00      0.00         3\n",
            "    Site Preparation_Site Security Setup       0.33      0.50      0.40         2\n",
            "                        Structural_Beams       1.00      1.00      1.00         3\n",
            "                      Structural_Columns       1.00      1.00      1.00         3\n",
            " Structural_Concrete Structure Formation       0.85      1.00      0.92        11\n",
            "                         Structural_Slab       0.89      0.89      0.89         9\n",
            "                         Structural_Wall       0.00      0.00      0.00         1\n",
            "                        Structural_Walls       1.00      0.67      0.80         6\n",
            "\n",
            "                                accuracy                           0.82       108\n",
            "                               macro avg       0.67      0.67      0.65       108\n",
            "                            weighted avg       0.81      0.82      0.80       108\n",
            "\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "c:\\Users\\PabloD\\anaconda3\\Lib\\site-packages\\sklearn\\metrics\\_classification.py:1344: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
            "  _warn_prf(average, modifier, msg_start, len(result))\n",
            "c:\\Users\\PabloD\\anaconda3\\Lib\\site-packages\\sklearn\\metrics\\_classification.py:1344: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
            "  _warn_prf(average, modifier, msg_start, len(result))\n",
            "c:\\Users\\PabloD\\anaconda3\\Lib\\site-packages\\sklearn\\metrics\\_classification.py:1344: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
            "  _warn_prf(average, modifier, msg_start, len(result))\n"
          ]
        }
      ],
      "source": [
        "import pandas as pd\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.feature_extraction.text import TfidfVectorizer\n",
        "from sklearn.svm import SVC\n",
        "from sklearn.metrics import classification_report\n",
        "from sklearn.pipeline import make_pipeline\n",
        "\n",
        "# Load training data\n",
        "\n",
        "df_train = pd.read_csv(training_data_path, usecols=['task_name', 'category', 'subcategory'])\n",
        "\n",
        "# Prepare training and target datasets\n",
        "X = df_train['task_name']\n",
        "y = df_train[['category', 'subcategory']].astype(str).agg('_'.join, axis=1)\n",
        "\n",
        "# Splitting the dataset for training and validation\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
        "\n",
        "# Creating a text vectorization and classification pipeline\n",
        "pipeline = make_pipeline(TfidfVectorizer(), SVC(kernel='linear', probability=True))\n",
        "\n",
        "# Training the model\n",
        "pipeline.fit(X_train, y_train)\n",
        "\n",
        "# Validating the model\n",
        "predictions = pipeline.predict(X_test)\n",
        "# Save the model for later use\n",
        "import joblib\n",
        "joblib.dump(pipeline, 'category_predictor.joblib')\n",
        "print(classification_report(y_test, predictions))\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3PgugLKrJY75"
      },
      "source": [
        "Once we have the model, we use it to clasiffy and store all the tasks we gather on the previous steps"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "q77hW9fhuO4o",
        "outputId": "79d4c681-97fc-473b-ad02-4bd08457d465"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Task categorization complete. Results have been saved.\n"
          ]
        }
      ],
      "source": [
        "\n",
        "# Load the file with tasks to classify\n",
        "classified_tasks_df = all_tasks_names\n",
        "assert 'task_name' in classified_tasks_df.columns, \"The CSV file must contain a 'task_name' column.\"\n",
        "\n",
        "# Predicting categories and subcategories for new tasks\n",
        "predicted_categories = pipeline.predict(classified_tasks_df['task_name'])\n",
        "classified_tasks_df['predicted'] = predicted_categories\n",
        "\n",
        "\n",
        "# Splitting the predicted values back into separate category and subcategory columns\n",
        "classified_tasks_df[['category', 'subcategory']] = classified_tasks_df['predicted'].str.split('_', expand=True)\n",
        "classified_tasks_df.drop(columns=['predicted'], inplace=True)\n",
        "\n",
        "# Saving the classified tasks back to the CSV\n",
        "classified_tasks_df.to_csv(classified_tasks_filepath, index=False)\n",
        "\n",
        "print(\"Task categorization complete. Results have been saved.\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "id": "U7irrJUXiYXX"
      },
      "outputs": [],
      "source": [
        "def join_dataframes_on_taskname(df1, df2):\n",
        "    # Ensure \"task_name\" is a column in both DataFrames\n",
        "    if \"task_name\" not in df1.columns or \"task_name\" not in df2.columns:\n",
        "        raise ValueError(\"Both dataframes must contain 'task_name' column\")\n",
        "\n",
        "    # Convert \"task_name\" in both DataFrames to uppercase for case-insensitive matching\n",
        "    df1_upper = df1.copy()\n",
        "    df2_upper = df2.copy()\n",
        "    df1_upper['task_name'] = df1_upper['task_name'].str.upper()\n",
        "    df2_upper['task_name'] = df2_upper['task_name'].str.upper()\n",
        "\n",
        "    # Perform the join operation\n",
        "    joined_df = pd.merge(df1_upper, df2_upper, on=\"task_name\", how=\"inner\")\n",
        "\n",
        "    return joined_df"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "id": "RyhWvoDUJY76"
      },
      "outputs": [],
      "source": [
        "joined_df = join_dataframes_on_taskname(final_df, classified_tasks_df)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {
        "id": "ozl4HDfMbwY9"
      },
      "outputs": [],
      "source": [
        "filtered_joined_df = joined_df[joined_df['category'] != 'Preliminary']\n",
        "# filtered_joined_df = joined_df\n",
        "filtered_joined_df.to_csv(\"joined.csv\", index=False)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nlSdFEm5lMvx"
      },
      "source": [
        "Complete merged dataset"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 12,
      "metadata": {
        "id": "ipTkt3-6lUJu"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "Unnamed: 0_x_x           127143\n",
              "proj_id_x                127143\n",
              "fy_start_month_num       133326\n",
              "chng_eff_cmp_pct_flag     98407\n",
              "rsrc_self_add_flag       133326\n",
              "                          ...  \n",
              "guid                          0\n",
              "ot_factor                     0\n",
              "rsrc_type                     0\n",
              "category                 133326\n",
              "subcategory              133326\n",
              "Length: 261, dtype: int64"
            ]
          },
          "execution_count": 12,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "filtered_joined_df.count()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 13,
      "metadata": {
        "id": "45xQnf6Hl4eB"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# @title Distribution of tasks by category and show totals\n",
        "\n",
        "from matplotlib import pyplot as plt\n",
        "import seaborn as sns\n",
        "filtered_joined_df.groupby('category').size().plot(kind='barh', color=sns.palettes.mpl_palette('Dark2'))\n",
        "plt.gca().spines[['top', 'right',]].set_visible(False)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 14,
      "metadata": {
        "id": "xqd4xBCyJY76"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "Index(['Unnamed: 0_x_x', 'proj_id_x', 'fy_start_month_num',\n",
              "       'chng_eff_cmp_pct_flag', 'rsrc_self_add_flag', 'allow_complete_flag',\n",
              "       'rsrc_multi_assign_flag', 'ts_rsrc_mark_act_finish_flag',\n",
              "       'ts_rsrc_vw_inact_actv_flag', 'checkout_flag',\n",
              "       ...\n",
              "       'location_id_y', 'location_id', 'Unnamed: 0', 'role_id', 'pobs_id',\n",
              "       'guid', 'ot_factor', 'rsrc_type', 'category', 'subcategory'],\n",
              "      dtype='object', length=261)"
            ]
          },
          "execution_count": 14,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "filtered_joined_df.columns"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "metadata": {
        "id": "tSkv1Ojz_8rT"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:5: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df.fillna(\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:5: FutureWarning: Downcasting object dtype arrays on .fillna, .ffill, .bfill is deprecated and will change in a future version. Call result.infer_objects(copy=False) instead. To opt-in to the future behavior, set `pd.set_option('future.no_silent_downcasting', True)`\n",
            "  filtered_joined_df.fillna(\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:5: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df.fillna(\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:12: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:14: FutureWarning: Downcasting behavior in `replace` is deprecated and will be removed in a future version. To retain the old behavior, explicitly call `result.infer_objects(copy=False)`. To opt-in to the future behavior, set `pd.set_option('future.no_silent_downcasting', True)`\n",
            "  filtered_joined_df.replace({\"N\": 0, \"Y\": 1}, inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:14: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df.replace({\"N\": 0, \"Y\": 1}, inplace=True)\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:32: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df['act_start_date'] = filtered_joined_df['act_start_date'].fillna(filtered_joined_df['target_start_date'])\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:35: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[date[0]] = (filtered_joined_df[date[1]] + (filtered_joined_df[date[2]] - filtered_joined_df[date[1]])/2  ).dt.month\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:36: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[date[0]] = filtered_joined_df[date[0]].astype(\"category\").cat.codes\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:35: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[date[0]] = (filtered_joined_df[date[1]] + (filtered_joined_df[date[2]] - filtered_joined_df[date[1]])/2  ).dt.month\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:36: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[date[0]] = filtered_joined_df[date[0]].astype(\"category\").cat.codes\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:35: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[date[0]] = (filtered_joined_df[date[1]] + (filtered_joined_df[date[2]] - filtered_joined_df[date[1]])/2  ).dt.month\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:36: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[date[0]] = filtered_joined_df[date[0]].astype(\"category\").cat.codes\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:35: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[date[0]] = (filtered_joined_df[date[1]] + (filtered_joined_df[date[2]] - filtered_joined_df[date[1]])/2  ).dt.month\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:36: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[date[0]] = filtered_joined_df[date[0]].astype(\"category\").cat.codes\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:43: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[date[0]] = (filtered_joined_df[date[1]] - filtered_joined_df[date[2]]).dt.days\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:43: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[date[0]] = (filtered_joined_df[date[1]] - filtered_joined_df[date[2]]).dt.days\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:43: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[date[0]] = (filtered_joined_df[date[1]] - filtered_joined_df[date[2]]).dt.days\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:43: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[date[0]] = (filtered_joined_df[date[1]] - filtered_joined_df[date[2]]).dt.days\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:46: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[\"delay\"] = filtered_joined_df[\"target_d\"] - filtered_joined_df[\"actual_d\"]\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:49: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[\"category\"] = filtered_joined_df[\"category\"].astype(\"category\").cat.codes\n",
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\3111129347.py:50: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  filtered_joined_df[\"subcategory\"] = filtered_joined_df[\"subcategory\"].astype(\"category\").cat.codes\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "              category  subcategory     delay  target_qty  successors  \\\n",
            "category      1.000000     0.291733 -0.008093    0.007710   -0.090442   \n",
            "subcategory   0.291733     1.000000 -0.001979   -0.006260   -0.031699   \n",
            "delay        -0.008093    -0.001979  1.000000    0.003370    0.048488   \n",
            "target_qty    0.007710    -0.006260  0.003370    1.000000   -0.001553   \n",
            "successors   -0.090442    -0.031699  0.048488   -0.001553    1.000000   \n",
            "actual_month  0.081906     0.077024 -0.073939   -0.004533   -0.041891   \n",
            "late_month    0.041998    -0.014491 -0.041953    0.004997   -0.072459   \n",
            "early_month   0.047637    -0.019903 -0.022136    0.012682   -0.006523   \n",
            "target_month  0.106107     0.089355 -0.332062   -0.008142   -0.094920   \n",
            "actual_d     -0.033460    -0.035164 -0.426148    0.001285    0.011588   \n",
            "late_d        0.130574     0.065506  0.266498   -0.009540   -0.047504   \n",
            "early_d       0.129871     0.063227  0.256696   -0.008739   -0.038155   \n",
            "target_d     -0.026607    -0.020761  0.878407    0.004398    0.059630   \n",
            "\n",
            "              actual_month  late_month  early_month  target_month  actual_d  \\\n",
            "category          0.081906    0.041998     0.047637      0.106107 -0.033460   \n",
            "subcategory       0.077024   -0.014491    -0.019903      0.089355 -0.035164   \n",
            "delay            -0.073939   -0.041953    -0.022136     -0.332062 -0.426148   \n",
            "target_qty       -0.004533    0.004997     0.012682     -0.008142  0.001285   \n",
            "successors       -0.041891   -0.072459    -0.006523     -0.094920  0.011588   \n",
            "actual_month      1.000000    0.037604     0.029187      0.512760 -0.262416   \n",
            "late_month        0.037604    1.000000     0.204919      0.137465 -0.005442   \n",
            "early_month       0.029187    0.204919     1.000000      0.119570  0.003610   \n",
            "target_month      0.512760    0.137465     0.119570      1.000000 -0.121864   \n",
            "actual_d         -0.262416   -0.005442     0.003610     -0.121864  1.000000   \n",
            "late_d            0.021091    0.036293     0.044212     -0.016406 -0.023994   \n",
            "early_d           0.025759    0.047066     0.035987     -0.010055 -0.023176   \n",
            "target_d         -0.220225   -0.049172    -0.022521     -0.430821  0.058014   \n",
            "\n",
            "                late_d   early_d  target_d  \n",
            "category      0.130574  0.129871 -0.026607  \n",
            "subcategory   0.065506  0.063227 -0.020761  \n",
            "delay         0.266498  0.256696  0.878407  \n",
            "target_qty   -0.009540 -0.008739  0.004398  \n",
            "successors   -0.047504 -0.038155  0.059630  \n",
            "actual_month  0.021091  0.025759 -0.220225  \n",
            "late_month    0.036293  0.047066 -0.049172  \n",
            "early_month   0.044212  0.035987 -0.022521  \n",
            "target_month -0.016406 -0.010055 -0.430821  \n",
            "actual_d     -0.023994 -0.023176  0.058014  \n",
            "late_d        1.000000  0.917184  0.281414  \n",
            "early_d       0.917184  1.000000  0.271029  \n",
            "target_d      0.281414  0.271029  1.000000  \n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1000x1000 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "\n",
        "# read the csv file classified_tasks.csv\n",
        "\n",
        "\n",
        "# fill the missing values with 0\n",
        "filtered_joined_df.fillna(\n",
        "    {col: 0 for col in filtered_joined_df.columns if filtered_joined_df[col].dtype != \"object\"}, inplace=True\n",
        ")\n",
        "\n",
        "# convert the date columns to datetime\n",
        "date_columns = [x for x in filtered_joined_df.columns if \"date\" in x]\n",
        "for col in date_columns:\n",
        "    filtered_joined_df[col] = pd.to_datetime(filtered_joined_df[col], errors=\"coerce\")\n",
        "# replace all N and Y with 1 and 0 respectively\n",
        "filtered_joined_df.replace({\"N\": 0, \"Y\": 1}, inplace=True)\n",
        "dates = [\n",
        "    [\"actual_d\", \"act_start_date\", \"act_end_date\"],\n",
        "    [\"late_d\", \"late_start_date\", \"late_end_date\"],\n",
        "    [\"early_d\", \"early_start_date\", \"early_end_date\"],\n",
        "    [\"target_d\", \"target_start_date\", \"target_end_date\"],\n",
        "\n",
        "]\n",
        "average_date = [\n",
        "    [\"actual_month\", \"act_start_date\", \"act_end_date\"],\n",
        "    [\"late_month\", \"late_start_date\", \"late_end_date\"],\n",
        "    [\"early_month\", \"early_start_date\", \"early_end_date\"],\n",
        "    [\"target_month\", \"target_start_date\", \"target_end_date\"],\n",
        "]\n",
        "\n",
        "columns = [\"category\", \"subcategory\",\"delay\", \"target_qty\", \"successors\" ]\n",
        "\n",
        "#fix missing dates columns, we will use\n",
        "filtered_joined_df['act_start_date'] = filtered_joined_df['act_start_date'].fillna(filtered_joined_df['target_start_date'])\n",
        "for date in average_date:\n",
        "    #get the average month between\n",
        "    filtered_joined_df[date[0]] = (filtered_joined_df[date[1]] + (filtered_joined_df[date[2]] - filtered_joined_df[date[1]])/2  ).dt.month\n",
        "    filtered_joined_df[date[0]] = filtered_joined_df[date[0]].astype(\"category\").cat.codes\n",
        "    columns.append(date[0])\n",
        "\n",
        "\n",
        "\n",
        "for date in dates:\n",
        "    #get the difference in days between the two dates\n",
        "    filtered_joined_df[date[0]] = (filtered_joined_df[date[1]] - filtered_joined_df[date[2]]).dt.days\n",
        "    columns.append(date[0])\n",
        "\n",
        "filtered_joined_df[\"delay\"] = filtered_joined_df[\"target_d\"] - filtered_joined_df[\"actual_d\"]\n",
        "\n",
        "# replace strings with integers for columns category and subcategory\n",
        "filtered_joined_df[\"category\"] = filtered_joined_df[\"category\"].astype(\"category\").cat.codes\n",
        "filtered_joined_df[\"subcategory\"] = filtered_joined_df[\"subcategory\"].astype(\"category\").cat.codes\n",
        "filtered_joined_df = filtered_joined_df[columns]\n",
        "# draw correlation matrix with matplotlib\n",
        "# use red color for negative correlation and blue color for positive correlation\n",
        "# a value of 1 means perfect positive correlation\n",
        "# a value of -1 means perfect negative correlation\n",
        "# use column name as labels for the x and y axis\n",
        "# x labeles to be vertical\n",
        "# add range to graphical representation\n",
        "plt.matshow(filtered_joined_df.corr(), cmap=\"coolwarm\", vmin=-1, vmax=1)\n",
        "\n",
        "plt.xticks(range(len(filtered_joined_df.columns)), filtered_joined_df.columns)\n",
        "plt.xticks(rotation=90)\n",
        "plt.yticks(range(len(filtered_joined_df.columns)), filtered_joined_df.columns)\n",
        "plt.colorbar()\n",
        "#make the plot bigger\n",
        "plt.gcf().set_size_inches(10, 10)\n",
        "# store it as a png file\n",
        "\n",
        "plt.savefig(\"correlation.png\")\n",
        "filtered_joined_df.to_csv(\"processed.csv\")\n",
        "p = filtered_joined_df.corr()\n",
        "p.to_csv(\"correlation.csv\")\n",
        "print(p)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 16,
      "metadata": {
        "id": "XDjiuM9AdfLK"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>category</th>\n",
              "      <th>subcategory</th>\n",
              "      <th>delay</th>\n",
              "      <th>target_qty</th>\n",
              "      <th>successors</th>\n",
              "      <th>actual_month</th>\n",
              "      <th>late_month</th>\n",
              "      <th>early_month</th>\n",
              "      <th>target_month</th>\n",
              "      <th>actual_d</th>\n",
              "      <th>late_d</th>\n",
              "      <th>early_d</th>\n",
              "      <th>target_d</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>39880</th>\n",
              "      <td>7</td>\n",
              "      <td>20</td>\n",
              "      <td>-21</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>4</td>\n",
              "      <td>7</td>\n",
              "      <td>7</td>\n",
              "      <td>7</td>\n",
              "      <td>0</td>\n",
              "      <td>-21</td>\n",
              "      <td>-21</td>\n",
              "      <td>-21</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>77123</th>\n",
              "      <td>7</td>\n",
              "      <td>2</td>\n",
              "      <td>-21</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0</td>\n",
              "      <td>10</td>\n",
              "      <td>10</td>\n",
              "      <td>10</td>\n",
              "      <td>0</td>\n",
              "      <td>-21</td>\n",
              "      <td>-21</td>\n",
              "      <td>-21</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>118502</th>\n",
              "      <td>6</td>\n",
              "      <td>16</td>\n",
              "      <td>-25</td>\n",
              "      <td>0.0</td>\n",
              "      <td>3.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>-25</td>\n",
              "      <td>-25</td>\n",
              "      <td>-25</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>118512</th>\n",
              "      <td>7</td>\n",
              "      <td>3</td>\n",
              "      <td>-30</td>\n",
              "      <td>0.0</td>\n",
              "      <td>5.0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>-30</td>\n",
              "      <td>-30</td>\n",
              "      <td>-30</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>118513</th>\n",
              "      <td>7</td>\n",
              "      <td>3</td>\n",
              "      <td>-30</td>\n",
              "      <td>0.0</td>\n",
              "      <td>5.0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>-30</td>\n",
              "      <td>-30</td>\n",
              "      <td>-30</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "        category  subcategory  delay  target_qty  successors  actual_month  \\\n",
              "39880          7           20    -21         0.0         1.0             4   \n",
              "77123          7            2    -21         0.0         1.0             0   \n",
              "118502         6           16    -25         0.0         3.0             0   \n",
              "118512         7            3    -30         0.0         5.0             0   \n",
              "118513         7            3    -30         0.0         5.0             0   \n",
              "\n",
              "        late_month  early_month  target_month  actual_d  late_d  early_d  \\\n",
              "39880            7            7             7         0     -21      -21   \n",
              "77123           10           10            10         0     -21      -21   \n",
              "118502           0            0             0         0     -25      -25   \n",
              "118512           3            3             3         0     -30      -30   \n",
              "118513           3            3             3         0     -30      -30   \n",
              "\n",
              "        target_d  \n",
              "39880        -21  \n",
              "77123        -21  \n",
              "118502       -25  \n",
              "118512       -30  \n",
              "118513       -30  "
            ]
          },
          "execution_count": 16,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "filtered_joined_df[filtered_joined_df[\"delay\"]< -20].head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 17,
      "metadata": {
        "id": "etiSLGZbF5Xz"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "-323\n",
            "122\n",
            "-0.7723099770487377\n"
          ]
        }
      ],
      "source": [
        "# prompt: print min max and mean of data['delay']\n",
        "\n",
        "print(filtered_joined_df['delay'].min())\n",
        "print(filtered_joined_df['delay'].max())\n",
        "print(filtered_joined_df['delay'].mean())"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 18,
      "metadata": {
        "id": "4IWwLnQzGIm1"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# prompt: Create a bar graph with bins of 25 for the values of data['delay']\n",
        "\n",
        "# Plot histogram with 1000 bins\n",
        "ax = filtered_joined_df['delay'].plot.hist(bins=1000, edgecolor='black')\n",
        "\n",
        "# Set x-axis limit to the min and max of the data\n",
        "ax.set_xlim(-10, filtered_joined_df['delay'].max())\n",
        "ax.set_ylim(0, 100)\n",
        "# Set y-axis limit from 0 to the max frequency in the bins\n",
        "\n",
        "\n",
        "# Set the title and labels\n",
        "ax.set_title('Histogram of Delays')\n",
        "ax.set_xlabel('Delay')\n",
        "ax.set_ylabel('Frequency')\n",
        "\n",
        "# Show the plot\n",
        "plt.show()\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "R2_GRgY9Pxtq"
      },
      "source": [
        "Approach checking the subcategories and the relation with delays"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 19,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Cr-Jf_f8ByUZ",
        "outputId": "312e4f7a-0931-48a9-89a6-0a1315cd1d94"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "C:\\Users\\PabloD\\AppData\\Local\\Temp\\ipykernel_32552\\1313628900.py:16: FutureWarning: A value is trying to be set on a copy of a DataFrame or Series through chained assignment using an inplace method.\n",
            "The behavior will change in pandas 3.0. This inplace method will never work because the intermediate object on which we are setting values always behaves as a copy.\n",
            "\n",
            "For example, when doing 'df[col].method(value, inplace=True)', try using 'df.method({col: value}, inplace=True)' or df[col] = df[col].method(value) instead, to perform the operation inplace on the original object.\n",
            "\n",
            "\n",
            "  df['target_qty'].fillna(df['target_qty'].mean(), inplace=True)\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            " Y max value: -1\n",
            "Accuracy: 0.991524788119703\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.82      0.97      0.89       917\n",
            "           1       1.00      0.99      1.00     25749\n",
            "\n",
            "    accuracy                           0.99     26666\n",
            "   macro avg       0.91      0.98      0.94     26666\n",
            "weighted avg       0.99      0.99      0.99     26666\n",
            "\n",
            " \n"
          ]
        },
        {
          "data": {
            "text/plain": [
              "['task_delay_classifier.joblib']"
            ]
          },
          "execution_count": 19,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "import pandas as pd\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.preprocessing import OneHotEncoder\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.metrics import classification_report, accuracy_score\n",
        "from sklearn.pipeline import Pipeline\n",
        "from sklearn.compose import ColumnTransformer\n",
        "\n",
        "# Load your data\n",
        "# Replace 'your_data.csv' with the path to your dataset\n",
        "df = filtered_joined_df\n",
        "\n",
        "\n",
        "# Or impute\n",
        "df['delay'].fillna(df['delay'].mean(), inplace=True)\n",
        "df['target_qty'].fillna(df['target_qty'].mean(), inplace=True)\n",
        "\n",
        "# Assume 'category' is the feature and 'late_d' is the target\n",
        "X = df[['category','subcategory',  \"target_month\", \"actual_month\", \"successors\" ]]  # Feature matrix\n",
        "ydef = df['delay']  # Target variable\n",
        "\n",
        "# Convert 'late_d' into a binary classification problem\n",
        "# Here, we define a delay as any task with 'late_d' > 0\n",
        "i = -1\n",
        "y = (ydef >= i).astype(int)\n",
        "\n",
        "# Split the dataset into training and testing sets\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=23)\n",
        "\n",
        "# Preprocessing: One-hot encoding for categorical variables\n",
        "preprocessor = ColumnTransformer(\n",
        "    transformers=[\n",
        "        ('cat', OneHotEncoder(handle_unknown='ignore'), ['category','subcategory', \"target_month\", \"actual_month\", \"successors\" ]),\n",
        "    ])\n",
        "\n",
        "# Create a pipeline with preprocessing and the model\n",
        "pipeline_delay = Pipeline(steps=[\n",
        "    ('preprocessor', preprocessor),\n",
        "    ('classifier', RandomForestClassifier(n_estimators=100, random_state=42, class_weight='balanced'))\n",
        "])\n",
        "\n",
        "# Train the model\n",
        "pipeline_delay.fit(X_train, y_train)\n",
        "\n",
        "# Predict on the test set\n",
        "y_pred = pipeline_delay.predict(X_test)\n",
        "\n",
        "# Evaluate the model\n",
        "print(f' Y max value: {i}')\n",
        "print(f'Accuracy: {accuracy_score(y_test, y_pred)}')\n",
        "print(classification_report(y_test, y_pred))\n",
        "print(f' ')\n",
        "\n",
        "# Save the model for later use\n",
        "import joblib\n",
        "joblib.dump(pipeline_delay, 'task_delay_classifier.joblib')\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ZZZ8_xVTrT6A"
      },
      "source": [
        "# Processing input\n",
        "\n",
        "Now that we have the model trained, we feed a real project into the process to detect the tipical delays"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "dR34ojJP1p-S"
      },
      "source": [
        "# Update a project\n",
        "\n",
        "Once we got valid results detecting delays we realize it was half way to a valid solution. We need to update a project with this information.\n",
        "\n",
        "We created a new model to detect and calculate delays.\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 20,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-iHsQciHz0L8",
        "outputId": "60ed1275-4c0d-499c-cff8-e955dfbf9693"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Average delay error - RMSE: 9.86834021881377\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1000x600 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "text/plain": [
              "['task_delay_regressor.joblib']"
            ]
          },
          "execution_count": 20,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.preprocessing import OneHotEncoder\n",
        "from sklearn.ensemble import RandomForestRegressor\n",
        "from sklearn.pipeline import Pipeline\n",
        "from sklearn.compose import ColumnTransformer\n",
        "from sklearn.metrics import mean_squared_error\n",
        "import numpy as np\n",
        "\n",
        "# Feature selection (Modify as per your dataset's columns)\n",
        "features = ['category', 'subcategory', 'target_month', 'successors']\n",
        "target = 'delay'  # This should be the actual observed delay days for each task\n",
        "\n",
        "# Convert categorical features to category types\n",
        "df['category'] = df['category'].astype('category')\n",
        "df['subcategory'] = df['subcategory'].astype('category')\n",
        "\n",
        "# Prepare the feature matrix X and target vector y\n",
        "X = df[features]\n",
        "y = df[target]\n",
        "\n",
        "# Split the data into training and testing sets\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
        "\n",
        "# Preprocessing for numerical and categorical data\n",
        "categorical_transformer = OneHotEncoder(handle_unknown='ignore')\n",
        "\n",
        "preprocessor = ColumnTransformer(\n",
        "    transformers=[\n",
        "        ('cat', categorical_transformer, ['category', 'subcategory']),\n",
        "    ])\n",
        "\n",
        "# Create a regression model pipeline\n",
        "regression_pipeline = Pipeline(steps=[\n",
        "    ('preprocessor', preprocessor),\n",
        "    ('regressor', RandomForestRegressor(n_estimators=100, random_state=42))\n",
        "])\n",
        "\n",
        "# Train the model\n",
        "regression_pipeline.fit(X_train, y_train)\n",
        "\n",
        "# Save the model for later use\n",
        "import joblib\n",
        "joblib.dump(regression_pipeline, 'task_delay_regressor.joblib')\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gYd0EBHC2BGF"
      },
      "source": [
        "# Process a Project\n",
        "\n",
        "\n",
        "Once we have the model trained we create a function to consume both models."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 21,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 504
        },
        "id": "GRYpIPJR1L7q",
        "outputId": "afea7fec-32fd-411e-bb92-47d24abb0731"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Updated task details saved to ./Baseline Zone 2 Rev/TASK_updated.csv\n"
          ]
        }
      ],
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "from datetime import timedelta\n",
        "import joblib\n",
        "\n",
        "# Load models\n",
        "pipeline_category = joblib.load('category_predictor.joblib')\n",
        "pipeline_delay_classifier = joblib.load('task_delay_classifier.joblib')\n",
        "pipeline_delay_regressor = joblib.load('task_delay_regressor.joblib')\n",
        "\n",
        "def add_successors(task_csv_path):\n",
        "    # Step 1: Load the tasks data\n",
        "    task_pred_path = os.path.dirname(task_csv_path) + \"TASKPRED.csv\"\n",
        "    task_df = pd.read_csv(task_csv_path)\n",
        "    if  os.path.exists(task_pred_path):\n",
        "        task_pred_df = fix_data(pd.read_csv(task_pred_path))\n",
        "        pred = task_pred_df.groupby('pred_task_id').agg(task_id_count=('task_id', 'count'))\n",
        "        pred.rename(columns={'task_id_count': 'successors'}, inplace=True)\n",
        "        pred = pred.rename_axis('task_id')\n",
        "        task_df = pd.merge(task_df, pred, on=\"task_id\", how=\"left\")\n",
        "    else:\n",
        "        task_df[\"successors\"] = 0\n",
        "    return task_df\n",
        "\n",
        "def update_task_details(task_csv_path, tasks_df):\n",
        "\n",
        "\n",
        "    # Step 1: Predict task categories and subcategories\n",
        "    if 'task_name' in tasks_df.columns:\n",
        "        categories_subcategories = pipeline_category.predict(tasks_df['task_name'])\n",
        "        tasks_df[['category', 'subcategory']] = pd.DataFrame([x.split('_') for x in categories_subcategories], index=tasks_df.index)\n",
        "\n",
        "    # Step 2: Preprocess data for delay prediction\n",
        "    tasks_df['category'] = tasks_df['category'].astype('category').cat.codes\n",
        "    tasks_df['subcategory'] = tasks_df['subcategory'].astype('category').cat.codes\n",
        "    tasks_df['target_month'] = pd.to_datetime(tasks_df['target_start_date']).dt.month\n",
        "    tasks_df['actual_month'] = pd.to_datetime(tasks_df['act_start_date'], errors='coerce').fillna(pd.to_datetime(tasks_df['target_start_date'])).dt.month\n",
        "    tasks_df['successors'] = tasks_df['successors'].fillna(0).astype(int)\n",
        "\n",
        "    # Prepare the feature matrix for delay prediction\n",
        "    features = tasks_df[['category', 'subcategory', 'target_month', 'actual_month', 'successors']]\n",
        "\n",
        "    # Step 3: Predict if there will be delays using the delay classification model\n",
        "    delay_predictions = pipeline_delay_classifier.predict(features)\n",
        "\n",
        "    # Step 4: Predict the number of delay days for tasks identified to likely have delays\n",
        "    delay_days = pipeline_delay_regressor.predict(features[delay_predictions == 1])\n",
        "\n",
        "    # Step 5: Update target_end_date only for delayed tasks\n",
        "    delay_indices = tasks_df.index[delay_predictions == 1]  # Indices of tasks predicted to be delayed\n",
        "    for idx, days in zip(delay_indices, delay_days):\n",
        "        if -(days) > 1:\n",
        "            # Assuming days is a variable containing the number of days to add (can be positive or negative)\n",
        "            tasks_df.at[idx, 'target_end_date'] = (pd.to_datetime(tasks_df.at[idx, 'target_end_date'], errors='coerce') + timedelta(days=int(days))).strftime('%Y-%m-%d %H:%M')\n",
        "\n",
        "    return tasks_df\n",
        "\n",
        "\n",
        "tasks_df = add_successors(Tasks_to_be_preditected)\n",
        "\n",
        "tasks_df = update_task_details(Tasks_to_be_preditected, tasks_df)\n",
        "# Optionally save the updated DataFrame to a new CSV file\n",
        "updated_csv_path = Tasks_to_be_preditected.replace('.csv', '_updated.csv')\n",
        "print(f\"Updated task details saved to {updated_csv_path}\")\n",
        "tasks_df.to_csv(updated_csv_path, index=False)\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "# Format\n",
        "Format the dataframe to CSV compatible with MS Project\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 22,
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "CSV has been formatted and saved for MS Project.\n",
            "CSV has been formatted and saved for MS Project.\n"
          ]
        }
      ],
      "source": [
        "def convert_to_MS_Project(task_csv, pred_csv, output_filename):\n",
        "    # Load the task data CSV file\n",
        "    task_df = pd.read_csv(task_csv)\n",
        "    # Load the predecessor data CSV\n",
        "    pred_df = pd.read_csv(pred_csv)\n",
        "\n",
        "\n",
        "    # Check if 'task_id' exists in both DataFrames\n",
        "    if 'task_id' not in task_df.columns:\n",
        "        raise ValueError(\"Column 'task_id' is missing in one of the files.\")\n",
        "    if 'task_id' not in pred_df.columns:\n",
        "        raise ValueError(\"Column 'task_id' is missing in one of the files.\")\n",
        "\n",
        "    # Merge predecessor data into the task data based on task ID\n",
        "    pred_df = pred_df[['pred_task_id', 'task_id']]\n",
        "\n",
        "    merged_df = task_df.merge(pred_df[['task_id', 'pred_task_id']], on='task_id', how='left')\n",
        "\n",
        "    # Rename and format columns as per MS Project requirements\n",
        "    ms_project_df = merged_df[['task_id', 'task_name', 'target_start_date', 'target_end_date', 'total_float_hr_cnt', 'pred_task_id']].copy()\n",
        "    # convert 'pred_task_id' to integer\n",
        "    ms_project_df['pred_task_id'] = ms_project_df['pred_task_id'].fillna(0).astype(int).replace(0, '')\n",
        "    # drop repeted rows\n",
        "    ms_project_df = ms_project_df.drop_duplicates(subset=['task_id'], keep='first')\n",
        "    #reset index\n",
        "    ms_project_df.reset_index(drop=True, inplace=True)\n",
        "    ms_project_df.rename(columns={\n",
        "        'task_id': 'Task ID',\n",
        "        'task_name': 'Task Name',\n",
        "        'target_start_date': 'Start Date',\n",
        "        'target_end_date': 'Finish Date',\n",
        "        'total_float_hr_cnt': 'Duration (h)',\n",
        "        'pred_task_id': 'Predecessor'\n",
        "    }, inplace=True)\n",
        "\n",
        "    # Format dates to match MS Project's expected format\n",
        "    ms_project_df['Start Date'] = pd.to_datetime(ms_project_df['Start Date'], format='%Y-%m-%d %H:%M').dt.strftime('%m/%d/%Y')\n",
        "    ms_project_df['Finish Date'] = pd.to_datetime(ms_project_df['Finish Date'], format='%Y-%m-%d %H:%M').dt.strftime('%m/%d/%Y')\n",
        "    \n",
        "    ms_project_df['Sorted Task ID'] = ms_project_df.index + 1\n",
        "    # using the Predecessor number, get task_id and replace Predecessor with the new task_id\n",
        "    for index, row in ms_project_df.iterrows():\n",
        "        if row['Predecessor'] != '':\n",
        "            task_id = ms_project_df.loc[ms_project_df['Task ID'] == row['Predecessor'], 'Sorted Task ID'].values\n",
        "            ms_project_df.at[index, 'Predecessor'] = task_id[0]\n",
        "    # Save the formatted CSV\n",
        "    ms_project_df.to_csv(output_filename, index=False)\n",
        "    print(\"CSV has been formatted and saved for MS Project.\")\n",
        "\n",
        "predecessors = Tasks_to_be_preditected.replace('TASK.csv', 'TASKPRED.csv')\n",
        "\n",
        "baseline_project = Tasks_to_be_preditected.replace('.csv', '_baseline_processed.csv')\n",
        "convert_to_MS_Project(Tasks_to_be_preditected, predecessors,baseline_project )\n",
        "\n",
        "new_file_name = Tasks_to_be_preditected.replace('.csv', '_updated_processed.csv')\n",
        "convert_to_MS_Project(updated_csv_path, predecessors, new_file_name)"
      ]
    }
  ],
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.11.7"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}